Re: [Intel-gfx] [PATCH] drm/i915/selftests: keep same cache settings as timeline

2023-03-17 Thread Matt Roper
On Thu, Mar 16, 2023 at 08:43:46PM -0700, Yang, Fei wrote: > >> From: Fei Yang > >> > >> On MTL, objects allocated through i915_gem_object_create_internal() are > >> mapped as uncached in GPU by default because HAS_LLC is false. However > >> in the live_hwsp_read selftest these watcher objects

Re: [Intel-gfx] [PATCH] drm/i915/selftests: keep same cache settings as timeline

2023-03-16 Thread Yang, Fei
>> From: Fei Yang >> >> On MTL, objects allocated through i915_gem_object_create_internal() are >> mapped as uncached in GPU by default because HAS_LLC is false. However >> in the live_hwsp_read selftest these watcher objects are mapped as WB >> on CPU side. The conseqence is that the updates

Re: [Intel-gfx] [PATCH] drm/i915/selftests: keep same cache settings as timeline

2023-03-16 Thread Matt Roper
On Wed, Mar 15, 2023 at 11:08:00AM -0700, fei.y...@intel.com wrote: > From: Fei Yang > > On MTL, objects allocated through i915_gem_object_create_internal() are > mapped as uncached in GPU by default because HAS_LLC is false. However > in the live_hwsp_read selftest these watcher objects are

[Intel-gfx] [PATCH] drm/i915/selftests: keep same cache settings as timeline

2023-03-15 Thread fei . yang
From: Fei Yang On MTL, objects allocated through i915_gem_object_create_internal() are mapped as uncached in GPU by default because HAS_LLC is false. However in the live_hwsp_read selftest these watcher objects are mapped as WB on CPU side. The conseqence is that the updates done by the GPU are

[Intel-gfx] [PATCH] drm/i915/selftests: keep same cache settings as timeline

2023-03-08 Thread fei . yang
From: Fei Yang On MTL, objects allocated through i915_gem_object_create_internal() are mapped as uncached in GPU by default because HAS_LLC is false. However in the live_hwsp_read selftest these watcher objects are mapped as WB on CPU side. The conseqence is that the updates done by the GPU are